The representer theorem for Hilbert spaces: a necessary and sufficient condition

Abstract

A family of regularization functionals is said to admit a linear representer theorem if every member of the family admits minimizers that lie in a fixed finite dimensional subspace. A recent characterization states that a general class of regularization functionals with differentiable regularizer admits a linear representer theorem if and only if the regularization term is a non-decreasing function of the norm. In this report, we improve over such result by replacing the differentiability assumption with lower semi-continuity and deriving a proof that is independent of the dimensionality of the space.

Cited by in corpus (5)

• Joint Ranking SVM and Binary Relevance with Robust Low-Rank Learning for Multi-Label Classification
• Learning from Distributions via Support Measure Machines
• Learning Output Kernels for Multi-Task Problems
• The Mathematical Foundations of Manifold Learning
• Geometric Interpretation of Running Nyström-Based Kernel Machines and Error Analysis